IHS Global Insight Perspective | |
Significance | The United Kingdom's National Institute for Health and Clinical Excellence (NICE) has said that Tyverb and Revolade are not cost-effective treatments for second-line, HER2-positive, advanced or metastatic breast cancer, and chronic immune thrombocytopenic purpura, respectively. |
Implications | Despite various modifications to the economic model for Tyverb and the addition of a patient access scheme in a bid to reduce costs, NICE maintains that the drug is too expensive in comparison with other treatments available on the market. NICE has identified various concerns in terms of the economic analysis for Revolade. |
Outlook | The guidance for Tyverb is final; however, GlaxoSmithKline (GSK) can still launch an appeal. At this stage, the changes of getting a recommendation appear to be slim. The best GSK can hope for is that local primary-care trusts will reimburse the drug despite this guidance, as the eligible patient population is small. The guidance for Revolade is preliminary, and GSK can now make revisions to the economic model based on the issues that NICE has raised. |
Final Guidance on Tyverb
U.K. drug-reimbursement watchdog the National Institute for Health and Clinical Excellence (NICE) has issued a final appraisal determination in which it notes that U.K. pharma company GlaxoSmithKline's (GSK's) Tyverb/Tykerb (lapatinib), in combination with capecitabine (Xeloda; Roche, Switzerland), is not recommended for the treatment of women with previously treated, HER2-positive, advanced or metastatic breast cancer.
In terms of the economic analysis of the drug, GSK used four main comparators: capecitabine monotherapy, vinorelbine monotherapy, Herceptin (trastuzumab; Roche; Switzerland) monotherapy, and combination therapy. The base-case incremental cost-effectiveness ratios (ICERs) per quality-adjusted life year (QALY) gained for Tyverb in combination with capecitabine were £81,251 in comparison with capecitabine monotherapy, and £67,847 in comparison with vinorelbine monotherapy. When compared with Herceptin regimens, Tyverb in combination with capecitabine was shown to be more effective and cheaper.
In the revised base-case analysis that took into account additional clinical data and data on the administration of Herceptin, the ICERs per QALY gained for lapatinib in combination with capecitabine were £93,825 when compared with capecitabine alone, £78,503 when compared with vinorelbine alone, and £24,227 when compared with Herceptin. GSK also introduced a blended comparator that represented a weighted average of the other comparators. This produced an ICER of £60,730 per QALY gained. When a patient-access scheme was introduced into the model, the ICERs per QALY gained were reduced to £16,384 for the blended comparator, £69,932 for the capecitabine comparator, and £54,610 for the vinorelbine comparator. Lapatinib plus capecitabine dominated all Herceptin regimens once again. Under the patient-access scheme, GSK agreed to pay for the first 12 weeks of treatment.
Based on these data, the committee concluded that when compared with capecitabine and vinorelbine monotherapy, lapatinib in combination with capecitabine is not a cost-effective use of National Health Service (NHS) resources. In terms of the comparison with Herceptin regimens, despite the fact that the economic model showed Tyverb in combination with capecitabine to be cheaper and more effective, the committee said that this was based on an indirect comparison of the two treatments, which it did not consider to be appropriate. In addition, the assumptions made in the model were highly sensitive to changes. The full appraisal is available here.
Preliminary Guidance on Revolade
In a separate preliminary recommendation, NICE said that another GSK drug, Revolade (eltrombopag), is not recommended for the treatment of chronic immune (idiopathic) thrombocytopenic purpura (ITP). This recommendation applies to patients whose condition is refractory to other treatments, and second-line treatment for non-splenectomised adults when surgery is contraindicated. Economic analysis was carried out for two treatment regimens.
- Watch-and-Rescue Treatment Model: This model studied the cost effectiveness of Revolade in combination with the standard of care for six months, in comparison with standard of care alone. This produced an ICER per QALY gained of £77,496 for the splenectomised subgroup, and £90,471 for the non- splenectomised subgroup.
- Long-Term Continuous-Treatment Model: This model evaluated the cost effectiveness of Revolade in patients with severe chronic ITP who do not respond to watch-and-rescue treatment. It was based on a sequence of treatments with Revolade, MabThera (rituximab; Roche, Switzerland), intravenous immunoglobulin, anti-D immunoglobulin, and Nplate (romiplostim; Amgen, U.S.). The ICERs per QALY gained ranged from £6 to over £545 million for the splenectomised subgroup, and from £6 to over £200 million for the non-splenectomised subgroup, depending on the order of treatments.
The committee concluded that Revolade is not a cost-effective use of NHS resources. The full appraisal is available here.
Outlook and Implications
This final decision on Tyverb comes as no surprise, as NICE has consistently maintained that the drug is too expensive, despite several attempts on GSK's part to demonstrate its cost effectiveness. The annual cost of Tyverb in combination with capecitabine is estimated to be £25,207. This decision comes despite the fact that the drug is reimbursed in various European countries, such as Poland, Croatia, and Italy (as part of a risk-share agreement; see Italy: 10 July 2009: GSK's Tykerb in Risk-Sharing Agreement with AIFA for Breast Cancer in Italy).
This decision means that up to 2,000 women who could possibly be treated with the drug may be denied treatment. However, the final decision on whether the drug should be reimbursed rests with individual primary-care trusts, as NICE's guidance is not binding. In contrast, rival drug Herceptin is reimbursed in the United Kingdom, and despite the fact that GSK's economic model shows that Tyverb in combination with capecitabine is potentially cheaper than Herceptin regimens, the lack of direct comparisons of the two dugs makes it difficult to provide concrete evidence to support this. GSK will have an opportunity to appeal this decision, but the chances of NICE overturning the final guidance appear to be slim (see United Kingdom: 22 October 2009: NICE Reverts to Original Negative Recommendation for GSK's Tyverb).
According to NICE, the negative preliminary decision on Revolade is driven by a number of concerns about both models. For the watch-and-rescue treatment model, these concerns hinged around the timeframe used in the model and several assumptions made in the model, as well as the utility values used. For the long-term continuous-treatment model, the main sticking points were the suitability of the clinical data used in the model, and the most appropriate treatment pathway. The appraisal is open for comments until 1 July, and a second consultation meeting will be held on 27 July.